Role of Deep Learning in Improving the Performance of Driver Fatigue Alert System
Bibliographic record
Abstract
Driver fatigue detection system aims to monitor the driver state. When detecting a fatigue caused by different attitudes other than normal driving habit, the system warns the driver that traveling should be interrupted. In this way, it helps the driver to make the right decision. The aim of this study is to prevent traffic accidents. The system analyzes any changes in the driver's eyes and mouth features in real time and warns the driver when necessary. The proposed system contains several stages to detect the driver's fatigue. First, the preprocessing stage; enhancement of the frames, determining the face, and cropping eyes and mouth of the driver was done. Then, dealing with feature extraction stage; the features concerning each frame was processed. Finally, two classification approaches were presented and a comparison between them was addressed. In the first approach, four traditional classifiers were applied; Diagonal Linear Discriminant Analysis (DiagLDA), Linear Support Vector Machine (LSVM), K-Nearest Neighbor (KNN), and Random Forest Classifier (RFC). The results show that two classifiers; KNN and RFC yield the highest average accuracy of 91.94% for all subjects presented in this paper. In the second approach, one model of deep learning neural network (CNN) was applied; "Resnet-50" model. The results also show that the proposed deep learning model yields a high average accuracy of 96.3889% for the same data. In general, the drowsiness and lost focus of drivers with high accuracy have been detected with the developed image processing based system, which makes it practicable and reliable for real-time applications.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".